An improved algorithm based on CLAHE for ultrasonic well logging image enhancement

Abstract

This study presents a novel image enhancement method in consideration of low contrast and blurred details of ultrasonic well logging images by combining the contrast limited adaptive histogram equalization (CLAHE) with power-law transformation. The red, green, and blue (RGB) color image is first transformed into hue, saturation, and intensity (HSI) space and the CLAHE is applied only to the intensity (or luminance) component (I) of the image. This enhances local details and prevents excess brightness in flat areas and uniform regions while maintaining the chromaticity values of the degraded image. Limitation of the CLAHE technique is that it limits the amplification of contrast by clipping the histogram of every sub-region at a predefined clip-limit. The power-law transformation is then applied to the processed image following the CLAHE processing. This maps a wide range of dark input values into a narrower range of output values, whereas a narrow range of bright input values into a wider range of output values, which further improve the contrast and highlight the local details. Finally, the result is presented as an RGB image. The performance of the proposed technique is compared with histogram equalization and CLAHE methods based on image quality objective measurement tools such as the mean square error (MSE) involving peak signal-to-noise ratio (PSNR), information entropy (IE), and luminance contrast (LC). Experimental results show that the proposed algorithm produces better ultrasonic well logging images than the existing enhancement schemes. The underlying improvement is of great significance for further upgrading the accuracy of sonic imaging logging data interpretation.